This study of diarrheal disease in infancy will define the frequency of disease due to enterotoxigenic E. coli. Specimens will be obtained from jejunal aspirates in selected infants and from stool or rectal swab specimens. Enterotoxigenicity will be determined by intragastric inoculation of newborn mice and confirmed in young rabbit intestinal loops. Enteroinvasiness will be tested in tissue culture using a Hep-2 cell line. Screening methods will be developed for rapid identification of the two types of E. coli and antibiotic treatment studies to determine the efficacy of orally non-absorbable antibiotics versus those which produce effective tissue concentrations. In addition, improved methods of identification of enterotoxin by gel diffusion, indirect hemagglutination or fluorescent antibody techniques will be attempted.